U.S. patent number 7,912,660 [Application Number 12/495,988] was granted by the patent office on 2011-03-22 for system and method for locating and analyzing arcing phenomena.
This patent grant is currently assigned to Exacter, Inc.. Invention is credited to Larry Anderson, John W. Beal, David S. Ellis, John Lauletta.
United States Patent |
7,912,660 |
Anderson , et al. |
March 22, 2011 |
System and method for locating and analyzing arcing phenomena
Abstract
System and method for detecting partial discharge arcing
phenomena in a power network distribution system which employs a
mobile receiving assemblage including a wideband antenna, a
computer controllable wideband radio receiver deriving an amplitude
detected output and a global positioning system providing system
position data. The amplitude detected outputs are digitized and
treated with a digital signal processor based analysis including
fast Fourier transforms extracting narrowband signal frequencies
that are harmonically related to the network fundamental frequency.
The narrowband signal frequencies are analyzed for peak amplitudes
which are summed to derive maintenance merit values related to the
arcing phenomena.
Inventors: |
Anderson; Larry (Columbus,
OH), Beal; John W. (Delaware, OH), Ellis; David S.
(Hilliard, OH), Lauletta; John (Hudson, OH) |
Assignee: |
Exacter, Inc. (Columbus,
OH)
|
Family
ID: |
38987427 |
Appl.
No.: |
12/495,988 |
Filed: |
July 1, 2009 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20090265122 A1 |
Oct 22, 2009 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
11881939 |
Jul 30, 2007 |
7577535 |
|
|
|
60905424 |
Mar 7, 2007 |
|
|
|
|
60834475 |
Jul 31, 2006 |
|
|
|
|
Current U.S.
Class: |
702/62; 702/61;
702/60; 340/636.17; 340/870.07; 362/536; 362/531; 340/870.16;
340/650; 340/646 |
Current CPC
Class: |
G01R
31/1227 (20130101); G01R 29/0814 (20130101) |
Current International
Class: |
G01R
21/00 (20060101); G08B 21/00 (20060101) |
Field of
Search: |
;702/60,61,62
;340/870.07,870.16,636.17,646,650 ;362/531,536 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Feliciano; Eliseo Ramos
Assistant Examiner: Suglo; Janet L
Attorney, Agent or Firm: Mueller Smith & Okuley, LLC
Parent Case Text
CROSS-REFERENCE TO RELATED APPLICATIONS
This application is a Continuation of application Ser. No.
11/881,939, filed on Jul. 30, 2007, and claims priority on
Provisional application Nos. 60/905,424 filed Mar. 7, 2007, and
60/834,475, filed Jul. 31, 2006, the disclosures of which are
expressly incorporated herein by reference.
Claims
We claim:
1. A method for detecting and generally locating arcing at a
component of an electrical power distribution or transmission
system, said system by means of radio frequency emissions,
exhibiting a fundamental a.c. frequency and located within a given
geographic region, comprising the steps: providing a mobile
receiving assemblage including a wideband antenna, a computer
controllable wideband radio receiver capable of receiving said
radio frequency emissions having an audio output, and a global
positioning system (GPS) receiver providing position data;
maneuvering said receiving assemblage about said geographic region;
converting said audio output to digital form at a sampling rate to
provide digital samples; providing a control assemblage including a
digital signal processor, a control computer, a fast Fourier
transform (FFT) function and a storage function; responding to said
digital samples with said FFT function of said control assemblage
to derive narrow band signal values that are harmonically related
to said fundamental a.c. frequency; and further with said control
assemblage, detecting said signal values and classifying them
according to arc strength to provide classified signal values,
controlling said wideband radio receiver with respect to its
detection range, deriving values of maintenance merit from said
classified signal values, correlating said values of maintenance
merit when greater than a set point, with position data from said
GPS receiver to provide position associated maintenance merit
values and submitting them to said storage function, further with
said control assemblage control computer, comparing a said
maintenance merit value with a low setting, when such maintenance
merit value is less than such low setting, then when radio receive
frequency charge capability is available, said control computer
effecting a lowering of the wideband radio receive frequency of
said radio receiver, said control computer being responsive when
such maintenance merit value is greater than said low setting to
effect, when radio receiver frequency change capability is
available, a raising of the wideband radio receive frequency of
said radio receiver.
2. The method of claim 1 in which: said control assemblage is
maneuverable with said receiving assemblage.
3. A system for detecting and generally locating an arcing
phenomenon within an electrical power distribution or transmission
system which exhibits wideband radio frequency emission with a
varying amplitude said varying amplitude changing at the electrical
power distribution system fundamental a.c. frequency and its
harmonics and said phenomenon being located within a given
geographic region, comprising: a mobile receiving assemblage
including a wideband antenna, a computer controllable wideband
radio receiver operatively coupled with said wideband antenna and
having an amplitude detected output, and a global positioning
system (GPS) receiver providing GPS position data; an
analog-to-digital (A/D) converter responsive to said amplitude
detected output to convert it to digital form at a sampling rate to
provide digital samples; a digital signal processor (DSP)
configured for carrying out arc detection and analysis including
fast Fourier transforms (FFT) of said digital samples, extracting
narrow band signal frequencies therefrom that are harmonically
related to said fundamental frequency, analyzing said harmonically
related narrow band frequencies for peak amplitudes and summing
such peak amplitudes to derive maintenance merit values; and a
control computer including a digital storage facility, radio
receive frequency change capability, said control computer being
responsive to control said radio receiver to locate said amplitude
detected output, said control being responsive to compare a said
maintenance merit value with a low setting, when such maintenance
merit value is less than such low setting, then when radio receive
frequency change capability is available said control computer
effecting a lowering of the wideband radio receive frequency of
said radio receiver, said control computer being responsive when
such maintenance merit value is greater than said low setting to
effect, when radio receiver frequency change capability is
available, a raising of the wideband radio receive frequency of
said radio receiver, said control computer further being responsive
to compile said maintenance merit values with said GPS position
data, and submit such compiled data to the storage facility.
4. The system of claim 3 in which: said analog-to-digital
converter, said digital signal processor, said control computer and
said radio receiver are mounted within a portable housing.
5. The system of claim 4 in which: said portable housing is located
within a vehicle maneuverable about said geographic region, said
wideband antenna and GPS receiver being mountable with said
vehicle.
6. The system of claim 5 further comprising: an arc data storage
server configured to receive said compiled merit value and GPS
position data from a cell phone network; and a cellular modem
within said housing controllable by said control computer to
broadcast said compiled merit values and position data to said arc
data storage server.
7. The system of claim 6 in which said system further comprises: a
display facility in data transfer communication with said arc data
storage server and configured to display a map of said geographic
region in combination with visible indicia representing said
compiled merit values and GPS position data and also displays GPS
track information indicating geographical coverage by the mobile
system.
8. The system of claim 5 in which: said vehicle includes a storage
battery electrical power supply and an ignition switch actuateable
between on and off orientations to selectively activate a switched
electrical power supply; said system further comprising a power
supply circuit under the control of said control computer, said
power supply circuit being responsive to actuation of the ignition
switch to the on-orientation to power all components of said system
from the switched electrical power supply and said power supply
circuit being controllable in response to actuation of said
ignition switch to the off-orientation to effect the powering of
the vehicle borne components of said system from the storage
battery electrical power supply for an interval effective to
broadcast said compiled merit value and position data.
9. The system of claim 3 in which: said digital signal processor is
further configured to carry out a finite impulse response filtering
of said maintenance merit values to provide resultant maintenance
merit values for said compilation with GPS position data.
10. The system of claim 9 in which: said control computer is
responsive to a said resultant merit value when it exceeds a
setpoint value and is derived in the presence of said fundamental
frequency or a harmonic thereof to submit said compiled data to the
storage facility.
11. The system of claim 3 in which said system further comprises: a
weather sensing assemblage having a weather output representing
ambient temperature, humidity and barometric pressure; and said
control computer is responsive to submit said weather output to
said storage facility in conjunction with the submittal of said
compiled merit values and GPS position data.
12. The system of claim 3 in which: said control computer sets said
analog-to-digital converter sample length to a mathematical power
of 2 and sample rate such that said fast Fourier transforms exhibit
an output with said narrow band frequencies falling on said
fundamental frequency and its harmonics.
13. The system of claim 3 in which: said control computer is
responsive to submit said digital samples as raw data to the
storage facility to develop a signature analysis capability.
14. The system of claim 3 further comprising: an arc data storage
server configured to receive said compiled data from a cell phone
network; and a cellular modem within said housing controllable by
said control computer to broadcast said compiled data to said arc
data storage server.
15. The system of claim 14 further comprising: a display facility
in data transfer communication with said arc data storage server
and configured to display a map of said geographic region in
combination with visible indicia representing said compiled
data.
16. The system of claim 3 in which: said control computer sets the
sample length of said first and second analog-to-digital converters
to derive a fast Fourier transform sequence length which is a
mathematical power of 2 and said fast Fourier transforms of said
first and second digital signal processors exhibit outputs with
said narrow band frequencies falling on said fundamental and its
harmonics.
17. A system for detecting and generally locating an arcing
phenomena within an electrical power distribution or transmission
system which exhibits wideband radio frequency emissions with a
varying amplitude changing at the electrical power distribution
system fundamental a.c. frequency and its harmonics and located
within a given geographic region, comprising: a mobile receiving
assemblage including at least one wideband antenna, at least one
computer controllable wideband radio receiver operatively coupled
with said wideband antenna and having an amplitude detected output,
and a global positioning system (GPS) receiver providing GPS
position data; an analog-to-digital (A/D) converter responsive to
said amplitude detected output to convert said amplitude detected
output to digital form at a sampling rate to provide digital
samples; a digital signal processor (DSP) configured for carrying
out arc detection and analysis including fast Fourier transforms
(FFT) of said digital samples extracting narrow band signal
frequencies from said fast Fourier transforms that are harmonically
related to said fundamental frequency, and analyzing said
harmonically related narrow band frequencies for peak amplitudes,
and summing such peak amplitudes to derive maintenance merit
values; a failure signature library having an input for receiving
detected and analyzed arc data including said fast Fourier
transforms of said digital samples, extracted narrow band signal
frequencies that are harmonically related to said fundamental
frequency, the peak amplitudes of a said fast Fourier transform
analysis extracting narrow band signal frequencies, a radio
frequency spectrum of a said analysis, an accept/reject signature
event indicator, a signature part type, a signature part number and
a manufacturer; a signature correlation and selection filter
controllable to correlate said failure signature library retained
arc data with the said carrying out of arc detection and analysis
prior to said analysis for peak amplitudes; and a control computer
including a digital storage facility and said control computer
being responsive to control said radio receiver to locate said
amplitude detected output, said control computer being responsive
to compare a said maintenance merit value with a low setting, when
such maintenance merit value is less than such low setting then
when radio receive frequency change capability is available said
control computer effecting a lowering of the wideband radio receive
frequency of said radio receiver, said control computer being
responsive when such maintenance merit value is greater than said
low setting to effect, when radio receiver frequency change
capability is available, a raising of the wideband radio receive
frequency of said radio receiver, said control computer further
being responsive to control said signature correlation and
selection filter to an extent effective to enhance the derivation
of maintenance merit values, to reject known false signature
indications and further responsive to compile said maintenance
merit values signature correlation data with said GPS position
data, and submit such compiled data to the storage facility.
18. The system of claim 17 in which: said analog-to-digital
converter, said digital signal processor, said control computer and
said radio receiver are mounted within a portable housing.
19. The system of claim 18 in which: said portable housing is
located within a vehicle maneuverable about said geographic region,
said wideband antenna and GPS receiver being mountable with said
vehicle.
20. The system of claim 19 further comprising: an arc data storage
server configured to receive said compiled merit value and GPS
position data from a cell phone network; and a cellular modem
within said housing controllable by said control computer to
broadcast said compiled merit values and position data to said arc
data storage server.
21. The system of claim 20 in which said cellular modem and said
failure signature library are controlled by said control computer
to download said detected and analyzed arc data to said failure
signature library input.
22. The system of claim 20 in which said system further comprises:
a display facility in data transfer communication with said arc
data storage server and configured to display a map of said
geographic region in combination with visible indicia representing
said compiled merit values and GPS position data.
23. The system of claim 19 in which: said vehicle includes a
storage battery electrical power supply and an ignition switch
actuateable between on and off orientations to selectively activate
a switched electrical power supply; said system further comprising
a power supply circuit under the control of said control computer
said power supply circuit being responsive to actuation of the
ignition switch to the on-orientation to power all components of
said system from the switched electrical power supply and said
power supply circuit being controllable in response to actuation of
said ignition switch to the off-orientation to effect the powering
of the vehicle borne components of said system from the storage
battery electrical power supply for an interval effective to
broadcast said compiled merit value and position data.
24. The system of claim 17 in which: said digital signal processor
is further configured to carry out a finite impulse response
filtering of said maintenance merit values to provide resultant
maintenance merit values for said compilation with GPS position
data.
25. The system of claim 24 in which: said control computer is
responsive to a said resultant merit value when it exceeds a
setpoint value and is derived in the presence of said fundamental
frequency or a harmonic thereof to submit said compiled data to the
storage facility.
26. The system of claim 17 in which said system further comprises:
a weather sensing assemblage having a weather output representing
ambient temperature, humidity and barometric pressure; and said
control computer is responsive to submit said weather output to
said storage facility in conjunction with the submittal of said
compiled merit values and GPS position data.
27. The system of claim 17 in which: said control computer sets
said analog-to-digital converter sample length to a mathematical
power of 2 and sample rate such that said fast Fourier transforms
exhibit an output with said narrow band frequencies falling on said
fundamental frequency and its harmonics.
28. The system of claim 17 in which: said control computer is
responsive to submit said digital samples as raw data to the
storage facility to develop a signature analysis capability.
Description
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH
Not applicable.
BACKGROUND
Electric utilities represent the largest energy provider to
consumers at the industrial, commercial and residential levels. The
infrastructure to support the delivery of energy to the 131 million
customers in the U.S. has been evolving over 100 years to what it
is today. During this time, the distribution voltage level has
become standardized at 69 kV, 34.5 kV, 17.2 kV, 12 kV, 7.2 kV, 4.7
kV, and 2.7 kV. These voltages are transformed to 480, 240, and 120
volts for use in machinery, systems, and homes.
The electrical distribution network can be compared to UPS van
delivery service. UPS vans do not manufacture the products they
deliver, nor do they carry them cross country. A UPS van is used to
distribute packages at a local level.
The electrical distribution system or network connects to the
electrical transmission system to deliver energy to the end-user.
The transmission system connects to the generation system where
electric energy is produced. Together, these three systems
instantly deliver energy to all customers on demand. However, if
the demand becomes too great or if the distribution system breaks
down, there is no alternative way to deliver energy to end-users
and a blackout or outage occurs.
The distribution system is different than the transmission system.
The transmission system is easily noticed along highways and in the
countryside. It includes imposing structures, long cross-country
transmission lines, and power plants. Conversely, the distribution
system is part of the urban landscape.
The distribution system or network is made up of 20- to 40-foot
wooden or steel poles from which are suspended power lines or
conductors, disconnect devices, lightning arrestors, capacitors,
insulators, and a variety of pole line hardware elements, each of
which plays a crucial role in keeping the lights on and factories
running. This uptime of the electrical system or network is defined
by one word: reliability.
The distribution system connects to the electric utility at a
substation where transmission voltages are reduced to distribution
levels. One transmission circuit delivers energy over many
distribution circuits. A substation is like a UPS freight terminal
where large cross-country trucks break down their loads to be
picked up and delivered by smaller UPS vans. Power lines supported
by power poles are referred to as overhead lines.
Due to deregulation in the 1990's, which segregated power
generation from the distribution of electricity, maintenance of
electric distribution systems is no longer fully included in a
utility's rate base. This has resulted in a 50% reduction in
electric distribution maintenance spending since 1990.
Electric utilities are largely regulated by state and federal
entities that monitor pricing, consumer satisfaction, and
reliability. The state PUCs are tasked with the regulation of
pricing and customer reviews of the electric utility as a
monopolistic supplier of energy. The PUC has the right and ability
to deny a utility the right to increase the charge for energy to a
class of customers based upon public hearings and customer review.
A recent rate hike for a major power company was denied in April of
2006 because according to the PUC of Ohio (PUCO) there was a
"failure to maintain baseline performance levels of 75% of its
distribution circuits." Baseline performance can be interpreted as
keeping the lights on enough of the time to avoid customer
complaints to the state PUC.
There are over three million miles of overhead and underground
electrical distribution circuits in the U.S. that provide consumers
access to electrical energy. Ninety percent of all interruptions to
electrical service occur when elements of the distribution system
break down.
In this report, particular attention is paid to the electrical
distribution network as an aging infrastructure that is being
continuously strained without the appropriate level of attention
and rehabilitation. See: www.energetics.com/gridworks/grid.html,
Department of Energy.
Decisions made by Public Utility Commissions (PUC) on the price a
utility can charge for energy are often affected by the costs that
consumers must bear for unreliable service. A study by the
Department of Energy (DOE) in 2004 found that there is an annual
$79 billion cost to consumers resulting from power outages.
Thirty-nine states have some form of punitive rate impact based
upon customer satisfaction and the number of outages within a
utilities service territory. These states each have mandatory
outage reporting requirements and reliability measurement
targets.
Some power failures like those due to natural disasters are
unavoidable, but avoidable outages from failure of circuit elements
or components make up 31% of all outages as measured by the
National Energy Regulatory Commission (NERC).
In 2003, there were $2.4 billion in electric rate cases pending
with PUC's. A rate case deferral or reduction of 54% or $1.3
billion was levied based upon service reliability and customer
satisfaction issues.
Electric utilities in the U.S. experience an estimated 6 million
outages each year related to electric distribution mechanical
failures. This has resulted in a loss of $750 million annually to
the utilities, in addition to failures to receive rate increases
due to unsatisfactory reliability performance amounting to as much
as $1.3 billion in 2003.
Thus, the electric utilities are dealing with the conflicting goals
of delivering strong financial performance for investors while
providing increasingly higher reliability performance for state
utility regulators.
The distribution system is the delivery point for all utility
customers, except the largest industrial customers such as steel
mills and automotive manufacturing plants. Large industrial
customers purchase energy on a wholesale basis at transmission
voltage levels of 138 kV, 230 kV, 345 kV, 500 kV, or 765 kV. These
wholesale customers include other utilities that buy and sell power
on the wholesale market.
Power is the instantaneous measure of energy. Energy is power
consumed over time and this is what small customers, like homes,
purchase. Power is measured in thousands of watts or kilowatts
(kW). Energy is how much power is used over time and is measured in
thousands of watts per hour of use, or kilowatt-hours (kWh). At
home we pay for energy by the kWh which averages about 10 cents per
kWh. A typical home load may consume 1,000 kWh per month or about
$100 of energy.
If a failure occurs in the distribution system, a customer looses
electrical service. At the same time, the utility is impacted in
several ways. First, it cannot sell energy to its customers.
Second, customers immediately complain to the utility and the
utility must react to these complaints. The utility must staff
complaint lines, pay overtime to repair crews, locate problems and
dispatch crews to the trouble area, purchase unplanned circuit
elements to replace those that have failed, and explain to the
Federal Energy Regulatory Commission (FERC) and the PUC why the
problem occurred, what it is doing to avoid the problem
reoccurring, and try to regain its customers' confidence through
public relations and advertising.
A power outage or history of unreliable service also raises the
issue of competition. In most states, electric customers have the
opportunity to select who will provide them their energy. This has
come about as part of the Energy Deregulation Act. This has set up
fierce competition between the largest Investor-Owned Utilities
(IOUs) like AEP, Con Ed and Duke Energy and 100 others; federal
utilities including the Bonneville Power Administration (BPA), the
Tennessee Valley Authority (TVA), and the Western Area Power
Administration (WAPA); the Rural Electric Cooperatives (COOP) of
which there are over 2,000; and the Municipal Electric Companies
like the City of Columbus in Columbus, Ohio, of which there are
thousands. Each of these entities must maintain customer loyalties
or risk customer migration.
The distribution circuit or system is supported by a number of
hardware elements. These elements maintain proper operation when
they are all working. Age, vibration, weather, air pollution,
lightning, and load all work against these elements causing them to
loosen, crack, and fail. As these elements or components begin to
fail, they emit high-frequency signals (EMI). These signals become
pronounced as the element nears catastrophic failure or flashover.
The result of a failure is an outage on the circuit feeding
thousands of customers.
When an energized component fails, there is a telltale emission
that results from electrical energy escaping from the circuit. This
is much like a radio antenna, broadcasting the imminent failure.
Devices have been designed that can report these emissions to
expensive computer-based communication networks. The basic
signature of failures is an arc which evidences an R.F. output
exhibiting a very steep rise time followed by a decay. Important
energy involved is one evoked from the rise time and not the decay.
Looking to some component failures, with a broken distribution
insulator, the electrical field surrounding the insulator begins to
leak through the broken areas of it and sharp edges of the fracture
emit these (EMI) signals that are detectable. The broken device
becomes critical with a flashover of the insulator and an outage of
the associated distribution circuit. Conductor brackets are
designed to hold an energized conductor in place and maintain
proper spacing from all other elements of the distribution system.
If such a bracket fails, the conductor becomes loose and could
swing into nearby structures of vegetation. If the conductor
contacts any structure, tree or other path to ground, an outage
occurs. Freeze-thaw cycles of weather may be a culprit in the
causation of a loose conductor bracket. Conductors themselves may
be partially broken from overload or other mechanical damage. The
broken strands of the conductor limit the loads that can be
supported before the conductor fails electrically. These strands
may also serve as small antenna which emits specific signals.
The distribution circuit or system is a single path for the
delivery of energy to homes, businesses, and industry. It begins at
the step-down transformer at a substation. The step-down
transformer reduces the voltage of the circuit from transmission
levels to lower distribution voltages. An involved conductor or
conductors in the entire network is energized to the distribution
voltage level until the distribution transformer reduces the
voltage once more to the appropriate low delivery voltage. A home
usually receives a voltage of 120 volts line to neutral or 240
volts line to line.
If any of the hardware connecting, insulating, or protecting the
distribution circuit or system fails, all of the loads downstream
of the failure become affected. Sometimes a power outage occurs
because there has been a problem such as a tree limb falling across
a line or an animal causing an electrical fault by bridging across
two conductors. However, equipment or component failure is the
leading cause of circuit failure. When an equipment or component
failure occurs, the broken element must be located and
replaced.
Power failure can be a nuisance to the homeowner. Who hasn't had to
reset their digital clocks following a power outage? But
long-duration outages--those outages resulting from equipment
failures--can cause serious damage particularly to a business which
relies on electricity to operate.
A national survey of 411 small-business operators conducted in
January 2004 by Decision Analysts for Emerson raises big questions
about the ability of small companies to withstand a lengthy power
outage. The survey, which is accurate to plus or minus five
percentage points, found that 80% of small businesses experienced
an electrical porter outage in 2003. Further is was determined that
60% have no type of back-up power supply. Also, a Small Business
Power Poll found that 75% of U.S. small businesses rate electrical
power outages as only marginally less of a threat than competition
(79%) and trauma from computer failure and a fire (77%). See:
Eckberg, John, "Power failures: Small companies, big losses," The
Cincinnati Enquirer, Mar. 14, 2004.
Weather plays a significant role in electrical distribution
equipment failure. When weather is inclement, a power outage is
more than a nuisance. In this regard, many Canadian home-heating
systems depend on electric power. Power lines and equipment can be
damaged by freezing rain, select storms, high winds, etc. This
damage can result in supply interruptions lasting from a few hours
to several days. An extended power failure during winter months and
subsequent loss of heat can result in cold, damp homes, severe
living conditions, and damage to walls, floors and plumbing.
Litigation resulting from power failures is often a secondary
effect. So much of the safety infrastructure on roadways, emergency
alert systems, and life-support systems are dependent upon reliable
energy.
Systems exist that address the concept of predictive circuit
review, but these systems require the problems to become so bad
that they are casually observed by customers. These are ultraviolet
(UV) and infrared (IR) imagery of the circuit elements. UV cameras,
such as those manufactured by OFIL of Israel, and IR cameras
manufactured by FLIR, Inc., are available.
Existing monitoring products have a relatively high base cost and
require technical skills, devoted labor, and post-analysis to be
effective. The effectiveness of these methods relies on the
opportunistic discovery of an already failing circuit element.
There is no discovery survey associated with their use.
BRIEF SUMMARY
The present disclosure is addressed to system and method for
locating and analyzing (e.g., partial discharge arcing phenomena),
as may be encountered in electrical power distribution networks and
the like. Those networks will perform at a given fundamental
frequency which in the United States, for example, will be 60 Hz or
25 Hz with respect to Amtrack. The arc detecting approach
incorporates one or more computer controllable wideband AM radio
receivers having an arc signal amplitude detected output. That
output is digitized to provide digital samples which are analyzed
with a digital signal processor utilizing fast Fourier transforms
to extract narrowband signal frequencies that are harmonically
related to the fundamental frequency of the network under
investigation. Such narrowband frequencies are further analyzed for
peak amplitudes which are summed to derive maintenance merit
values. A control computer is responsive to control the one or more
radio receivers to locate the amplitude detected output and compile
maintenance merit values with global positioning system data for
submittal to storage. With the system, displays or maps of arcing
phenomena may be published, the maintenance merit values giving an
indication of the intensity and thus the criticality of arcing
phenomena.
The system and approach is compact and does not require the
intervention of a technician to operate within a given geographical
area. In a preferred arrangement, the system is hardwired into the
battery power supply and ignition switch function of a vehicle
within which it rides. With such an arrangement, the system is
turned on in conjunction with actuation of the vehicle ignition
switch from an off position to an on position. When the vehicle
completes a journey within the given geographical region and the
ignition switch is turned off, the system will retain battery power
supply until it uploads all collected data to a server or the like
utilizing a cellular modem within a cell telephone system.
In one embodiment, the system employs two computer controllable
wideband radio receivers, a first being dedicated to high frequency
values of arcing phenomena and the second looking to lower
frequency phenomena. The lower frequency based radio is computer
adjusted based upon computed maintenance merit values, while the
higher frequency dedicated radio is adjusted by adjusting a look-up
table based upon the lower frequency maintenance merit values and
radio frequency response.
In still another approach to the system, arcing phenomena
characteristics are further analyzed utilizing a failure signature
library storing analyzed arc data including fast Fourier transforms
of digital sample, extracted narrowband signal frequencies
harmonically related to the network fundamental frequency, peak
amplitudes of such an analysis, a radio frequency spectrum of that
analysis, an accept/reject signature event indicator, a signature
part type, a signature part number and a manufacturer.
Other objects of the disclosure of embodiments will, in part, be
obvious and will, in part, appear hereinafter.
The instant presentation, accordingly, comprises embodiments of the
system and method possessing the construction, combination of
elements, arrangement of parts and steps which are exemplified in
the following detailed disclosure.
For a fuller understanding of the nature and objects herein
involved, reference should be made to the
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a perspective view of the instant system showing a
carrying case in a closed orientation;
FIG. 2 is a side view of the carrying case shown in FIG. 1;
FIG. 3 is a block diagram of the present system;
FIG. 4 is a plan view of the carrying case of FIG. 1 showing it in
an open orientation;
FIG. 5 is a perspective view of a distribution network insulator
component which is defective;
FIG. 6 is a block diagram of the instant system;
FIG. 7 is a schematic representation of a map produced by the
instant system;
FIG. 8 is a block diagram of the instant system showing its
incorporation of weather condition data;
FIG. 9 is a symbolic diagram of the instant system;
FIG. 10 is a symbolic representation of the instant system;
FIG. 11 is a block diagram of a power supply approach to the
instant system;
FIG. 12 is an electrical schematic representation of portions of
the power control circuit of FIG. 11;
FIG. 13 is a software block diagram of a single radio embodiment of
the present system;
FIG. 14 is a block diagram of the software associated with FFT and
harmonic strength calculations as represented in FIG. 13;
FIG. 15 is a block diagrammatic representation of peak harmonic
detection as described in connection with FIG. 13;
FIG. 16 is a block diagrammatic representation of maintenance merit
calculation as described in connection with FIG. 13;
FIG. 17 is a block diagrammatic representation of an FIR filter
function described in connection with FIG. 13;
FIG. 18 is a block diagrammatic representation of a recording
control function described in connection with FIG. 13;
FIG. 19 is an arc proximity computation block diagram as described
in connection with FIG. 13;
FIG. 20 is a software block diagram of a dual radio implementation
of the instant system;
FIG. 21 is a block diagram of an arc proximity computation approach
described in connection with FIG. 20;
FIG. 22 is a software block diagram of a single radio with
signature analysis embodiment of the instant system;
FIG. 23 is a block diagram of an FFT, harmonic strength and
correlation calculation described in connection with FIG. 22;
and
FIG. 24 is a block diagram of a signature correlation and selection
filter described in connection with FIG. 22.
DETAILED DESCRIPTION OF THE INVENTION
A salient feature of the present method and system resides in its
portability coupled with a capability of performing "on its own"
without the manual intervention of a technician. This carries to
the extent that when transported by a vehicle within a desired
geographic region, it turns itself on in conjunction with operator
actuation of a vehicle ignition switch to an on position and
uploads its retrieved and treated arc differentiation and location
data, for example, to an arc server when that switch is turned off.
The vehicle operator may drive a random or pre-designated route
within the subject geographic region. While somewhat technically
complex, the convenience of the system is manifested by its high
level of portability.
Looking to FIG. 1, such portability is made evident for the system
as represented in general at 10. System 10 principally is concerned
with a portable vehicle carried unit, the housing represented
generally at 12 of which is a polymeric industrial carrying case
with a handle 14, top and bottom components 16 and 18 which are
mutually hinged together and retained in a closed orientation by
over-center latches as at 20 and 22. Shown coupled to the housing
12 is a vehicle power input cable 24. Cable 24 preferably is
hardwired into the vehicle, for example, at a fuse box. Adjacent
cable 24 is cable 26 which extends to the systems cellular antenna
28. Antenna 28, for example, may be affixed to the cab roof of the
vehicle by virtue of its magnetic base. Next adjacent cable 26 is a
cable 30 extending from housing 12 to a global positioning system
receiver arrangement 32 which also may be provided with a magnetic
base for coupling to a vehicle roof. Finally, a cable 34 extends
from housing 12 to a wideband radio antenna 36 which also may be
provided with a magnetic base for purposes of vehicle roof
mounting. Shown additionally at bottom component 18 of housing 12
is a fan vent 38.
Looking momentarily to FIG. 2, the pivoting or hinged connection
between components 16 and 18 is shown generally at 40. Above this
connection 40 is a power input connector 42 which receives both
switched power and battery power. Adjacent the connector 42 is a
mobile antenna coupling 44. A cooling fan is provided at 46. Above
cooling fan 46 is a global positioning system (GPS) connector 48
and above that connector is the coupling for a wideband radio
receiver antenna 50. An audio out connector is shown at 51.
Looking to FIG. 3, a broad block diagrammatic representation of
system 10 is presented. In the figure, symbolically, wideband radio
antenna 36 reappears along with GPS receiver 32 and cellular
antenna 28. Cable 34, extending from antenna 36 again reappears as
a line extending to the wideband antenna input of a wideband radio
receiver represented in general at 52. Receiver 52 may be provided,
for example, as a IC-PCR 1500 communications receiver marketed by
Icom America, having a web location at
http://www.icomamerica.com/product/receivers/per1500/specs.asp.
Receiver 52 is controllable by computer as represented at line 54
extending from a control computer represented at block 56. Computer
56 may be provided as an Ampro Ready System.TM. 2 U Computer,
marketed by Ampro Computers, Inc. of San Jose, Calif. Note that the
audio output from wideband radio receiver 52 is represented at
arrow 58 extending to control computer 56. Also extending to the
control computer 56 is cable 30 of the GPS receiver identified
earlier at 32. Adjacent to that input is the cellular antenna
assemblage including antenna 28, cable 26 and a cellular modem
represented at block 60. The association between cellular modem 60
and control computer 56 is represented at line 62. Cellular modem
60 may be a type MTCBA-C marketed by Multi-Tech Systems, Inc., of
Mounds View Minn.
System 10 also incorporates a power control circuit represented
generally at block 64. Circuit 64 is associated with control
computer 56 by providing a shutdown command as represented at line
66 and being monitored by control computer 56 as represented at
line 68. Finally, system 10 may incorporate an audio-out feature as
represented at line 70, arrow 72 and dashed block 74 representing
FM voice transmission for providing prompts and the like,
particularly with respect to carrying out diagnostics.
Looking to FIG. 4, the interior of housing 12 is revealed,
components heretofore described being identified with the same
numeration. Additionally, a circuit association of components is
revealed somewhat in general by dashed lines. In this regard, a
dashed line 80 is seen extending between the connector for a
wideband receiving antenna 36 and wideband radio receiver 52. A
dashed line representing audio-out is shown at 82 extending from
connector 51 to control computer 56. GPS input to the control
computer 56 is shown at dashed line 84. Power is shown supplied to
fan 46 as represented at dashed line 86. Vehicle power-in earlier
described at connector 42 in connection with FIG. 2 is shown
coupled with power control board 64 as represented at dashed line
88. Connection between cellular modem 60 and control computer 56 is
shown at dashed line 90. Power-in to control computer 56 is
represented at dashed line 92, while power monitoring is
represented at dashed line 94. A communication of modem 60 with
earlier-described connector 44 (FIG. 2) is represented at line
dashed line 96.
Arcing may occur in connection with a broad variety of components
or structures within a given power distribution network. The arc is
characterized in having a waveform which very rapidly rises and
then decays. This creates a radio frequency interference condition
which often is a precursor to breakdown of regions of a
distribution network. Arcing and subsequent breakdown can occur in
conjunction with a broad variety of network components. In this
regard, an exemplar of a failed component is represented in FIG. 5
wherein there is pictorially represented in general at 102 a broken
distribution insulator. Such insulators as at 102 may, for example,
support a 13.5 kV conductor. Note that damage is represented in
general at 104. With such damage, the electric field surrounding
the insulator will begin to leak through the broken region and the
partial discharge near sharp edges of the fracture will be observed
to emit specific signals (EMI) that are detectable. Left
uncorrected, the end result would be a flashover of the insulator
and an outage of the distribution circuit. A detailed review of
such component defects is provided in the following publication:
Loftness, "A.C. Power Interference Handbook", Second Edition
Revised, 2003, Percival Technology, Tumwater Wash. 98501.
In a general context, the present system and method is represented
in connection with FIG. 6. Looking to that figure, arcing phenomena
(EMI) is represented at symbol 110 which is sensed as represented
at symbol 112 then, as represented at arrow 114 and block 116,
arcing phenomena are analyzed with respect to global position in
combination with a significance of an emission as translated into a
multi-dimensional parameter referred to as "maintenance merit".
Maintenance merit is a measure of the significance of an emission
from an arc source that includes evaluations such as: R.F. emission
spectrum; narrowband emission strength; demodulated narrowband
discharge emission spectrum; narrowband discharge emission
signature; fundamental and second harmonic detection (typically
25/50 Hz (Amtrack); 50/100 (Europe); or 60/120 Hz (USA)). The
parameter further incorporates detected signal temporal
information. With these components, the significance level of an
arc may be detected such that it may be prioritized. Next, as
represented at arrow 118 and block 120, the position related
maintenance merit data is transmitted via cellular modem for
processing by a server as represented at arrow 122 and block 124.
Once so processed, a map of the region of interest is produced or
displayed as represented at arrow 126 and block 128. In general,
the map will identify locations of arcing along with maintenance
merit level indication, for instance, with a color coding scheme.
Such a map is schematically represented in FIG. 7. Looking to that
figure, maintenance merit levels are identified symbolically, for
example, a highest level is represented at darkened squares 130. A
next lower level of priority is represented at darkened dots 132. A
third lower level of maintenance merit is represented at open
squares 134; and a lowest level of maintenance merit is represented
at open dots 136.
Referring to FIG. 8, a next level of detail of the instant system
is revealed. Power is derived from the transporting vehicle as
represented at block 140 and arrow 142. The receiving assemblage of
the system is represented at block 144 and the antenna symbol 146.
Received data is provided as an amplitude detected output as
represented at line 148 which is directed to an embedded computer
150. Additionally introduced to the control computer function at
block 150 is global positioning system data as represented at block
152 and line 154. Because arcing phenomena are influenced by
weather conditions, as represented at circles 156-158 and
respective lines 160-162, temperature, humidity and barometric
pressure data are directed as represented at block 164 and line 166
to the control computer 150. Finally, global positioning,
maintenance merit and weather data are generated and transmitted
through a cellular network to a processing function as described in
connection with FIG. 6 at 124 and herein shown as dual arrow 168,
block 170 and antenna symbol 172.
Referring to FIG. 9, a generalized representation of system 10 is
presented. In the figure, centrally disposed is a collector
function represented at symbol 180. A variety of data is collected
at function 180. In this regard, a wideband antenna is represented
at block 182 and arrow 184 as associated with a wideband amplitude
detect radio shown at symbol 186. The association of the radio
function 186 and collector function 180 is represented at line 188.
Temperature input to the collector function 180 is represented at
symbol 190 and line 192. Humidity data is introduced to collector
function 180 as represented at symbol 194 and line 196 and,
similarly, barometric pressure data is also submitted as
represented at symbol 198 and line 200. Global positioning system
information is represented at symbol 202. Note that symbol 202 is
represented as associated with an appropriate antenna as
represented at block 204 and arrow 206. Additionally, the symbol
202 function is associated with Greenwich Mean Time and date data
as represented at symbol 208 and line 210.
System power control is represented at symbol 212 seen associated
with the collector function 180 as represented at arrow 214. As
noted above, the power input to the power supply function 212 is
provided from an associated vehicle battery as represented at
symbol 216 and arrow 218. That power input is logically controlled
from the ignition switch of the vehicle as represented at symbol
220 and dual arrow 222.
The data collector function 180 is interactively associated with an
evaluation or evaluator function as represented at symbol 224 and
interactive arrows 226 and 228. The evaluation function 224 may
perform in conjunction with an arcing phenomena signature library
as represented at symbol 230 and interactive arrows 232 and 234. A
general storage function is represented at symbol 236 along with
interactive arrows 238 and 240. The cellular modem based up-loader
function is represented at symbol 242 along with interactive arrows
244 and 246. In general, when the vehicle ignition switch is turned
to an off position, uploading takes place. Where such uploading is
not successful, the system 10 will carry out a retry, repeating
three times as represented by loop arrow 248. As represented at
dual arrow 250 and symbol 252 uploading as well as downloading
takes place in conjunction with a web information portal or
server.
System 10 is further represented in connection with FIG. 10.
Looking to that figure, block 260 and associated R.F. emission
symbol 262 are represented as interacting with a wideband antenna
function represented at symbol 264. Wideband antenna function 264
performs in conjunction with a computer controlled wideband
programmable AM radio as represented at block 266. The function at
block 266 provides an amplitude detected arc signal as represented
at arrow 268 to an analog-to-digital converter function represented
at block 270. Function 270 provides digital samples as represented
at arrow 272 to a function represented at block 274. At block 274,
a digital signal processor configured for carrying out arc
detection and analysis is provided including fast Fourier
transforms of the digital samples, extracting narrowband signal
frequencies therefrom that are harmonically related to the
fundamental frequency of the distribution system, analyzing the
harmonically related narrowband frequencies for peak amplitudes and
summing such peak amplitudes to derive maintenance merit values.
Function 274 further includes a control computer which functions,
inter alia, to provide a radio frequency control over the wideband
AM radio function 266 as represented at arrow 276. Further provided
to the function 274 is global positioning system data as
represented at symbol 276 and arrow 278. Weather or environmental
data additionally is made available to the function 274 as
represented at block 280 and arrow 282. In carrying out its arc
analysis, the function 274 may perform in conjunction with a
failure signature library as represented at symbol 284 and arrow
286. Retained within this function 284 are detected and analyzed
arc data including fast Fourier transforms of digital samples,
extracted narrowband signal frequencies that are harmonically
related to the fundamental frequency of the network, the peak
amplitudes of such analysis, a radio frequency spectrum of the
analysis, an accept/reject signature event indicator, a signature
part type, a signature part number and associated manufacturer.
Maintenance merit values or arc strength are submitted to computer
storage as represented at symbol 288 and arrow 290. Power to the
entire system 10 is provided as above described and is represented
in the instant figure at block 292 and arrow 294.
Returning to storage function 288, with the actuation of an
associated vehicle ignition switch to an off orientation computer
controlled uplifting takes place utilizing a cellular modem as
represented at symbol 296 and arrow 298. As represented at arrow
300, cellular modem function 296 also performs a feature of adding
data to the failure signature library function 284. An uploading of
data by the cellular modem function 296 also functions to broadcast
via a cell phone network as represented at dual arrow 302 and
symbol 304. The cell phone network 304 additionally functions to
interact with an arc storage server as represented at block 306 and
dual arrow 308. Dual arrow 308 represents a feature wherein the
function 274 may be upgraded from a remote server location. Arc
storage server function 306 performs in conjunction with the
internet as represented at symbol 310 and interactive arrow 312.
The internet communicates arc event strength location and display
to a web portal display function as represented at block 314 and
dual arrow 316. Display function 314 may, for example, publish a
map as described in connection with FIG. 7.
Powering function of system 10 has been discussed, for example, in
connection with FIG. 3. Looking to FIG. 11, a more detailed
rendition of the power utilization feature of system 10 is
represented. In the figure, vehicle ground is represented at symbol
320 in conjunction with lines 322 and 324 extending to a connector
represented at block 326. Also extending to connector 326 is a +12
volt d.c. switched power input represented at symbol 328. Symbol
328 is shown associated with a one amp fuse function as represented
at symbol 330 and line 332. Connection with connector 326 from the
fuse function 322 is represented at line 334. +12 volt un-switched
vehicle power is represented at symbol 336. This un-switched power
function is represented as being directed through a 10 amp fuse as
shown at symbol 338 and line 340, whereupon connection with the
connector 326 is represented at line 342.
Connector 326 couples to connector 42 as described in FIG. 2 which
is represented in block form with the same numeration at the dashed
boundary representing housing 12. Connector 42 provides vehicle
ground as represented at lines 344 and 346 as well as the
earlier-described 12 volt d.c. switched power input at line 348 and
un-switched power input as represented at line 350. Lines 344, 346,
348 and 350 extend to corresponding inputs of power control circuit
64. Function 64 is under the control of the control computer
represented again at block 56. In this regard, power monitor arrows
68 reappears in conjunction with power-in or shutdown command arrow
66. 12 volts d.c. is provided to control computer function 56 as
represented at arrow 352 and ground is similarly supplied as
represented at arrow 354. The wideband radio receiver function
earlier-described at 52 in connection with FIG. 3 is shown
receiving 12 volts d.c. as represented at arrow 356 and
corresponding ground as represented at arrow 358. Similarly,
cooling fan 46 receives 12 volts d.c. as represented at arrow 360
and ground as represented at arrow 362.
Referring to FIG. 12, the power control circuit function
earlier-described at 64 is illustrated at an enhanced level of
detail. In the figure, the +vehicle battery input is provided at
line 370 while corresponding negative battery connection is
represented at line 372 extending to ground at line 374. Line 370
incorporates a 10 amp fuse F1 and the battery input is filtered for
noise control purposes by capacitors C1-C3 and inductor L1. The
filtered output of the battery input is presented to the Vcc
terminal of a solid state switch 376 as represented at line 378.
Switch function 376 may be, for example, a type IR 3314 and its
current control is provided by resistor R1 located within line 380
between device 376 and ground. The output of switch 376 is
represented in general at 382. This output is repeated on the
control board five times. Switch 376 is turned off and on by a
field effect transistor Q1, the drain of which is coupled via line
384 to device 376 and the source of which is coupled via line 386
to ground. Transistor Q1 is turned on either by the actuation of a
vehicle ignition switch to an on position or by an output of the
control computer of the system. The control computer function 56
signal produces an output shortly after the control computer power
382 is applied by actuation of the vehicle ignition switch to the
on position. The ignition switch signal at line 388, incorporating
resistor R2 and a form of steering diode D1 is coupled to the gate
of transistor Q1. The input at line 388 is divided down by a
network incorporating resistors R7 and R8 and a filtering capacitor
C6. Line 390 incorporating resistor R9 extends to a terminal 392
representing an input to the control computer corresponding with
line 66 described in connection with FIG. 3. Thus, the computer
function is provided a signal representing that ignition has been
turned off. A line 394 also extends to terminal 392. Upon receiving
a signal that the vehicle ignition has been turned off, the
computer control function 56 continues to provide an output at
lines 396 and 398 extended to line 388 to keep transistor Q1 on
until the cellular modem data upload function (FIG. 10, Block 296)
is complete, at which time the computer control function 56 signal
is removed turning off transistor Q1. This control computer serving
network as represented at 400 incorporates a steering diode D2,
divider resistors R3 and R4 and a filtering capacitor C4. Line 396
corresponds with earlier-described line 68 in FIG. 3.
Additional regulator networks may be provided in conjunction with
the output of device 376. In this regard, note that the output of
that device additionally is coupled to a network represented
generally at 402 via line 404. Network 402 includes a regulator 406
which may, for example be a type LM 317T with an input at line 408
and an output at line 410. Device 406 is configured with resistors
R5 and R6 as well as capacitor C5. Ground is provided from lines
412 and 414.
In the discussion to follow, block diagrams are presented
describing the software activity of system 10. Three approaches are
described, one involving a single AM radio function; one involving
two such radio functions; and the third describing a single radio
with a signature analysis feature. The blocks and symbols making up
the block diagrams have been provided using the SDL-2000
standardized specification and description language.
Looking to FIG. 13, a general block diagram of the single radio
embodiment is set forth. A wideband antenna is represented at
symbol 440 which is operationally associated with a computer
controllable wideband radio receiver represented at block 442. The
amplitude detected output of radio 442 will be between 0 and 6 kHz
as represented at arrow 444. The amplitude detected output then is
converted to digital form by an analog-to-digital (A/D) converter
function represented at block 446. Sampling rate derived digital
samples then are available as represented at arrow 448 and symbol
450. Such digital data is made available to raw data storage as
represented by arrow 452 and symbol 454. Just above symbol 454 is
symbol 456 and arrow 458 providing for the setting up of parameters
for all blocks of the diagram. Returning to symbol 450, as
represented at arrow 460 and block 462, the digital samples are
submitted to a fast Fourier transform (FFT) and harmonic strength
calculation function. Note that block 462 addresses the conversion
block 446 via arrow 464 to provide for sampling rate control.
Now considering the FFT, the width of one frequency bin in an FFT
can be calculated as "Sample Rate SR/FFT Sequence Length".
Therefore, to provide an exact 60 Hz bin and to fulfill a
requirement of two calculations per second, the optimal combination
is "Sample Rate=2*FFT Sequence Length". The frequency bin
resolution is then 2.0 Hz and the FFT bins are at 2, 4, 6, . . . ,
60, . . . , 120 Hz, etc.
In order to perform an FFT in a fast and efficient way, the FFT
Sequence Length must be a power of 2. To be able to extract exact
FFT values for desired frequencies and to provide desired data
rate, predefined values for sampling rate and FFT length are used.
The following is a tabulation of sampling rates in samples per
second; FFT length (samples) and number of FFTs in one second for
three frequencies, 60 Hz (USA); 50 Hz (Europe); and 25 Hz
(Amtrack).
TABLE-US-00001 Sampling FFT length Number of FFTs Frequency (Hz)
rate(samples/sec) (samples) in one second 60 32768 16384 2 50 32768
16384 2 25 40960 16384 2.5
After the tabulations the following parameters are adjustable and
set in system 10: 1. Number of channels is 1 or 2 (one for each
radio) 2. Main power line frequency (25, 50 or 60 Hz) 3. Number of
harmonics to calculate (1-100) 4. Filter Width in number of FFT
values to be taken into account (0 means exact frequency value, 1
means exact value and its left/right neighbors, etc.). The output
value for a single harmonic is calculated as:
.times. ##EQU00001## Where: FW=Filter Width Hi=Harmonic magnitude
value where (i) is harmonic number in the FFT Fi=FFT magnitude
value where (i) is the value index in the FFT
For optimum power line arc discrimination against other noise
sources the filter width should be as narrow as possible. Since the
filter width is inversely proportional to the FFT Sequence Length,
a longer length (and sample time) can be chosen to improve arc
signal discrimination.
Now looking momentarily to FIG. 14, block 462 is further diagrammed
to provide greater detail. Looking to that figure, symbol 470
represents the start of the main and transfer software thread. In
the latter regard, as represented at block 472 the transfer thread
carries out an initialization of the harmonic array buffer.
Returning to block 470, as represented at arrow 474 and block 476,
the analog-to-digital conversion rate is set. Next, as represented
at arrow 478 and block 480, one FFT is initialized and as
represented at arrow 482 and block 484 digital samples from the A/D
conversion process are read. This collection procedure continues as
represented at arrow 486 and symbol 488 determining whether the
collection procedure is done for this FFT. In the event that it is
not completed, the procedure continues as represented a loop arrow
490. Where data collection is completed, then as represented at
arrow 492 and block 494, the fast Fourier transform is performed.
Following performance of the FFT, as represented at arrow 496 and
block 498, a harmonic array of harmonically related scalar values
is computed and, as represented at arrow 500 and symbol 502, data
ready is set for the transfer thread (block 472) and the procedure
loops to initiate a next FFT as represented at loop arrow 504.
Returning to block 472, as represented at arrow 506 and symbol 508,
the procedure awaits the data ready condition as was set at block
502. When the data is ready, as represented at arrow 510 and block
512, the data is moved into a buffer and, as represented at arrow
514 and symbol 516, the display and peak harmonic detector are
alerted, whereupon the procedure loops to symbol 508 as represented
at arrow 518.
Returning to FIG. 13, with the completion of FFT and harmonic
strength calculation as represented at block 462, as indicated at
arrow 520 and block 522, the system analyzes the harmonically
related narrowband frequencies for peak amplitudes.
Looking momentarily to FIG. 15, this peak detection feature is
diagramed at an enhanced level of detail. In the figure, the
procedure commences in conjunction with start symbol 524 and arrow
526 extending to symbol 528. Symbol 528 provides for the awaiting
of a harmonic buffer-ready signal which, for example, will be
developed from symbol 516 described in connection with FIG. 14.
With the presence of a harmonic buffer-ready indication, as
represented at arrow 534 and block 536, a maximum value is found
and as represented at arrow 538 and block 540, that max harmonic
peak amplitude is saved. Next, as represented at arrow 542 and
symbol 544, a determination is made as to whether H0, which is the
power network fundamental frequency is greatest. If it is not, then
as represented at arrow 546 and symbol 548, a determination is made
as to whether the max amplitude is at a first harmonic of the
fundamental frequency. Where the query posed at either of symbols
544 or 548 results in an affirmative determination, then as
represented at either arrow 550 or arrow 552, the fundamental or
harmonic flag is set as indicated at symbol 554. In the event of a
negative determination at symbol 548, then as represented at arrow
556 and symbol 558, the procedure returns to FIG. 13 and arrow 560.
In considering the setting of the flags at symbol 554, temporal
information about an arc phenomena becomes available as a very
basic signature of the instant system. For instance, if a max
amplitude is associated with just the positive or the negative
going components of an assumed sinewave, then H0 flag is set. On
the other hand, where such amplitude is seen on both positive and
negative going components of the waveform, then H1 is set.
Now returning to FIG. 13, arrow 560 is seen directed to block 562
calling for the computation of a maintenance merit value. In
general, this is developed by summing the above-noted peak
amplitudes. Looking additionally to FIG. 16, this maintenance merit
computation feature is diagramed in more detail. The function is
shown entered as represented at symbol 570 and arrow 572 extending
to block 574. Block 574 calls for the summing of all harmonics,
whereupon, as represented at arrow 576 and block 578, an average of
the summed harmonics is computed. Because all harmonics are
sub-scale, as represented at arrow 580 and block 582, the computed
average is scaled to full scale and, as represented at arrow 584
and symbol 586, the scaled average is saved as a current
maintenance merit value. However, while usable with the system,
this value may be filtered. The procedure represented by block 562
then returns as represented at arrow 588 and symbol 590.
Accordingly, returning to FIG. 13, an arrow 592 is seen extending
from block 562 to block 594 calling for the filtering of the
maintenance merit computation employing a finite impulse response
(FIR) filter. Such filters also are referred to as averaging
filters and function to discriminate against noise. In general, the
user will determine a filtering length of maintenance merit values,
for example, up to 20. Looking to FIG. 17, the filtering function
of block 594 is diagramed at a higher level of detail. In the
figure, the filtering procedure is seen to commence in conjunction
with symbol 600 and arrow 602. Arrow 602 leads to symbol 604
providing for saving the current maintenance merit values in the
filter array, a limit of such values being elected for filtering.
Next, as represented at arrow 606 and block 608, filtering is
carried out by computing the sum of the maintenance merit values
for, n, such values divided by the limit value. Upon carrying out
this computational filtering, then as represented at arrow 610 and
symbol 612 the filtered maintenance merit (MM) value is saved. As
represented at arrow 614 and block 616, the index is increased by
one. Next, as represented at arrow 618 and symbol 620 a
determination is made as to whether the index (n) is greater than
the limit value minus one. In the event that it is, then as
represented at arrow 622 and block 624, the index, n, is set to
zero and as represented at arrows 626, 628 and symbol 629, the
procedure returns to block 594 in FIG. 13. As represented by arrow
628 and symbol. 629, the procedure reverts to block 594 shown in
FIG. 13. Returning to that block, arrow 632 is seen to extend
therefrom to symbol 634 showing that the maintenance merit value
now is a resultant one in consequence of the FIR filtering. System
procedure then continues as represented at arrow 640 and block 642
providing for recording control. In that regard, note that an arrow
644 extends to storage facility 454 as an indication that recording
is to be started. It may be recalled that this is raw data for
signature analysis.
Referring to FIG. 18, the recording control function 642 is
revealed at a higher level of detail. In the figure, the function
642 is entered as represented at symbol 650 and arrow 652 extending
to symbol 654 posing a query as to whether the maintenance merit
value is greater than a pre-selected setpoint. If it is not
greater, then that maintenance merit value is not recorded and the
procedure continues as represented at line 656 and exit symbol 658.
On the other hand, where the query at symbol 654 indicates that the
instant maintenance merit value is greater than a setpoint, then as
represented at arrow 660 and symbol 662, a determination is made as
to whether the H0 or H1 flag is set. It may be recalled that H0
represents power line fundamental frequency, while H1 represents a
first harmonic thereof. If neither of those flags is set, then the
data is neither recorded nor utilized and the procedure continues
as represented at arrows 664, 656 and symbol 658. On the other
hand, where either of those flags is set, then as represented at
arrow 666 and block 668, the system reads mean Greenwich time, GPS
location, temperature, humidity and barometric pressure. As
represented at arrow 670 and symbol 672, all such data is saved as
an arc event and function 642 provides a signal to start recording
as represented at arrow 674 and symbol 676. The procedure then
reverts to arrow 656 and symbol 658 as represented at arrow
678.
Returning to FIG. 13, as represented at arrow 680 and block 682,
the system carries out an arc proximity computation implementing
the computer controlled alteration of the frequency response of
radio 442. This association is represented at arrow 684. Function
682 changes the frequency response at radio function 442 based upon
the fingerprints being received. In this regard, if strong signals
are being received, a higher radiofrequency response will be
desired. This follows because of the nature of the arc signals
encountered, the higher the radiofrequency of such signals more
than likely the shorter the distance the system is from the arc
phenomena. On the other hand, arc signal phenomena travels longer
distances at lower radiofrequencies. Accordingly, an opposite form
of frequency response adjustment may be called for. Looking to FIG.
19, arc proximity computation function 682 is illustrated at a
higher level of detail. Function 682 is entered as represented at
symbol 686 and arrow 688 which is directed to the query posed at
symbol 690 determining whether the current maintenance merit value
is less than a low setting. If that is the case, then as
represented at arrow 692 and symbol 694, a determination is made as
to whether the RF frequency already is set at a low frequency
regime. If it has not been so set, then as represented at arrow 696
and block 698 the frequency response of radio function 442 is set
lower and, as represented at arrows 700, 702 and symbol 704, the
system returns to the peak harmonic detector function represented
in FIG. 13 at 522. That same result obtains if the query posed at
symbol 694 indicates that the RF frequency already has been set
low. With such a setting the system reverts to peak harmonic
detector function 522 (FIG. 13) as represented at arrows 706, 702
and symbol 704.
Returning to the query posed at symbol 690, where the current
maintenance merit value is not less than a low setting, then, as
represented at arrow 708 the system looks to the query at symbol
710 determining whether or not the RF frequency is at a maximum
level. In the event that it is at that maximum level, the system
again reverts to peak harmonic detector function 522 as represented
at arrow 702 and symbol 704. On the other hand, where the RF
frequency is not at a maximum level, then as represented at arrow
712 and block 714 the computer raises the frequency response of the
radio function 442 and, as represented at arrows 716, 702 and
symbol 704, the system reverts again to the function at block 522
in FIG. 13.
Improved arc phenomena detection and localization can be realized
by employing the system 10 with two wideband computer controllable
AM radios instead of one. Such a system is represented in general
at 720 at the block diagram presented in conjunction with FIG. 20.
In the figure, system 720 is seen to incorporate two wideband AM
radios 722 and 724 performing in conjunction with respective
antennae 726 and 728. The radiofrequency response of radios 722 and
724 again is computer controllable as represented at respective
arrows 736 and 732 extending from an arc proximity computation
function represented at block 734. As before, each of the radios
722 and 724 perform in conjunction with an amplitude detect output
which, for example, may be in the range of 0-6 kHz. For
convenience, computer controllable wideband radio receiver 722 is
referred to herein as radio no. 1 and its amplitude detected output
at arrow 736 is referred to as a first amplitude detected output.
In similar fashion, computer controllable wideband radio receiver
724 is referred to herein as radio receiver no. 2 and its amplitude
detected output is represented at arrow 738.
As in the case of FIG. 13 the block diagram of system 720 includes
a setup parameters symbol 750, and arrow 752 from which indicates
this feature applies to all blocks.
The first amplitude detected output as represented at arrow 736
from radio 722 is subjected to analog-to-digital conversion as
represented at block 754. As before, this conversion is rate
controlled as represented at arrow 756 and the output of this
conversion as represented at arrow 758 provides what is designated
herein as first high frequency parameter digital sample,
representing digital data from the first radio as indicated by
symbol 760.
In similar fashion, the output of radio no. 2 at arrow 738 may be
designated as a second amplitude detected output which also is
subjected to analog-to-digital conversion as represented at block
762. The sampling rate of converter 762 is computer controlled as
represented at arrow 764 and its output as represented at arrow 766
is herein designated as second low frequency parameter digital
sample represented, as shown in symbol 768 as digital data from
radio no. 2. The signal data from radio no. 1 as represented at
symbol 760, as in the case of FIG. 13, may be submitted as raw data
for signature analysis to storage or memory as represented at arrow
770 and symbol 772. However, as represented at arrow 774 and block
776, it is also now subjected to fast Fourier transform activity
and harmonic strength calculation. This is the same treatment as
described at block 462 in FIG. 13 as well as in connection with
FIG. 14. In similar fashion, the second low frequency parameter
digital samples as represented at symbol 768 are treated with a
fast Fourier transform and harmonic strength calculation as
represented at arrow 778 and block 780. As before, this function is
the same as that carried out in conjunction with block 462 in FIG.
13 and as described in FIG. 14.
Returning to block 776, a first digital signal processor has been
provided which is configured for carrying out arc detection and
analysis including fast Fourier transforms of the first digital
samples, extracting narrowband signal frequencies, (bins) that are
harmonically related to the fundamental frequency. Next, as
represented at arrow 782 and block 784, the harmonically related
narrowband frequencies are analyzed for peak amplitudes in a manner
identical to that described in connection with block 522 of FIG. 13
and the discourse presented in connection with FIG. 15.
Turning back to block 780, a second digital signal processor is
described which is configured for carrying out arc detection and
analysis including fast Fourier transforms of the second digital
samples, extracting narrowband signal frequencies therefrom (bins)
that are harmonically related to the fundamental frequencies and as
with radio 1, as represented at arrow 786 and block 788, analysis
is carried out of the harmonically related narrowband frequencies
for peak amplitudes in the same manner as described in connection
with block 522 of FIG. 13 and corresponding FIG. 15.
An arrow 790 extends from block 784 to block 792 providing for
maintenance merit computation in the same manner as described at
block 562 in connection with FIG. 13 and as further described in
connection with FIG. 16. Such maintenance merit values will be
identified in the instant figure as "MM1". In this regard, as
represented at arrow 794 and block 796, finite impulse response
filtering is carried out in the same manner as described in
conjunction with block 594 of FIG. 13 and as discussed in
connection with FIG. 17. A maintenance merit resultant, MM1 thus is
evolved as represented at arrow 798 and symbol 800.
Returning to the second radio component of the instant diagram, as
represented at arrow 802 and block 804, a maintenance merit
computation is carried out in the manner described in connection
with block 562 of FIG. 13 and as described in connection with FIG.
16. Then, as represented at arrow 806 and block 808, the
maintenance merit values are filtered utilizing a finite impulse
response filter in the manner described at block 594 in FIG. 13 and
as discussed in connection with FIG. 17. A result, as represented
at arrow 810 and symbol 800 is a resultant maintenance merit,
MM2.
Next, as represented at arrow 812 and block 814, the recording
control function is carried out in the manner described in
connection with block 642 of FIG. 13 and as described in more
detail in FIG. 18. Where the maintenance merit resultants are above
a setpoint, they are correlated with Greenwich mean time, GPS
location, temperature, humidity and pressure and recordation is
started as represented at arrow 816. Next, arc proximity
computation is carried out as represented at arrow 820 and block
734. For the instant embodiment utilizing radio no. 1 and radio no.
2, the arc proximity computation is somewhat altered, initially
looking to an analysis of the low frequency parameter maintenance
merit, MM2 and then doing a table look-up to set the high frequency
radio no. 1 frequency. Referring to FIG. 21, this altered approach
is diagramed in detail. In the figure, this feature is approached
as represented at symbol 822 and arrow 824 which is directed to the
query posed at symbol 826 determining whether maintenance merit MM2
(low frequency) is less than a low setting. In the event that it is
not less than a low setting, then as represented at arrow 828 and
symbol 830, a determination is made as to whether radio no. 2
frequency is at a maximum level. In the event that it is not, then
as represented at arrow 832 and block 834, the frequency setting at
radio no. 2 is raised and the program continues as represented at
arrow 836. If the adjustment of radio no. 2, (block 724) is at a
maximum setting, then the program continues as represented at block
838.
Returning to symbol 826 where the current maintenance merit (MM2)
is less than the low setting, then as represented at arrow 840 and
symbol 842, a query is posed as to whether radio no. 2 (RF2)
frequency already has been set at a low level. In the event that it
has not, then as represented at arrow 844 and block 846, the radio
no. 2 (RF2) frequency setting is lowered and the program continues
as represented at arrow 848 which extends to arrow 838. Returning
to symbol 842, where the radio no. 2 frequency setting is already
at a low level, then as represented at arrows 850, 848 and 838, the
program continues.
Arrow 838 is directed to block 852 which indicates that the
wideband radio frequency response ranges of the first computer
controllable radio receiver are retained in a look-up table
addressable by a combination of the second low frequency parameter
maintenance merit values and second wideband radiofrequency
response. Upon carrying out such look-up, as represented at arrow
854 and block 856, the radio frequency of radio no. 1 is set and
the program continues as represented at arrow 858 and symbol 860 to
reenter this dual program at blocks 784 and 788 as discussed in
connection with FIG. 20.
Another approach to the instant system involves the features of
FIG. 13 and system 10 as they are enhanced with a failure signature
library performing in conjunction with a signature correlation and
selection filter. In this regard, it may be recalled from FIG. 10
that new signatures were delivered to a failure signature library
from the cellular modem function. Looking to FIG. 22, this system
enhancement is represented generally at 870. In the figure, a
wideband antenna 872 is shown in operative association with a
computer controllable radio receiver represented at block 874. Such
computer control is over radio function 874 is represented at arrow
876. The amplitude detected output (0-6 kHz) from radio facility
874 is represented at arrow 878 which is directed to
analog-to-digital conversion as represented at block 880. Sample
rate control for the conversion function 880 is represented by
arrow 882. Also carried out is a set up of parameters as
represented at symbol 884, such set up applying to all blocks of
the diagram as represented at arrow 886. Returning to the
conversion function 880, digital samples are produced as
represented at arrow 888 to provide digital data from the radio
function 874 as represented at symbol 890. Arrow 892 represents
that such digital data is available to raw data storage as
represented at symbol 894.
Returning to symbol 890, as represented at arrow 896 and block 898,
digital signal processor is provided which is configured for
carrying out arc detection and analysis including fast Fourier
transforms (FFT) of the digital samples, extracting narrowband
signal frequencies (bins) therefrom that are harmonically related
to the fundamental frequency of the network.
Referring to FIG. 23, the function of block 898 is revealed at an
enhanced level of detail. Starting of the main software thread is
represented at symbol 902, while the transfer thread function
carries out an initialization of the harmonic array buffer as
represented at block 904. From symbol 902, an arrow 906 extends to
block 908 providing for setting the analog-to-digital conversion
rates a function represented in FIG. 22 at arrow 882. Next, as
represented at arrow 910 and block 912, one FFT is initialized and,
as represented at arrow 914 and block 916, the digital sample based
A/D data is read. As represented at arrow 918, symbol 920 and loop
arrow 922, such reading continues until the digital sample
collection is completed whereupon, as represented at arrow 924 and
block 926, the fast Fourier transform (FFT) is performed. Upon
completion of the FFT, as represented at arrow 928 and block 930,
the system computes a harmonic array of scalar values in the manner
described in connection with FIG. 14 and block 498. Upon such
computation, as represented at arrow 932 and symbol 934, data ready
is set for the transfer thread and, as represented at return arrow
936, the program reverts to block 912.
Returning to block 904, as represented at arrow 938 and block 940,
for the instant embodiment, comparison signatures are initialized
and as represented at arrow 942 and symbol 944, the system transfer
thread awaits a data ready input whereupon, as represented at arrow
946 and block 948, data is moved into a buffer and as represented
at arrow 950 and symbol 952, the system will communicate with a
signature correlation and selection filter shown in FIG. 22. As
represented at symbol 954, arrow 956 and symbol 958, a display and
peak harmonic detector is alerted and the thread loops as
represented by loop arrow 960 to symbol 944 awaiting another data
ready input.
Returning to FIG. 22, the failure signature library for this
enhancement is represented at symbol 962. It may be recalled from
FIG. 10 that these signatures are uploaded, inter alia, to this
library from the cellular modem function 296. Accordingly, in FIG.
22, downloading is represented at symbol 970 and arrow 972. Failure
signature library 962 receives and stores analyzed arc data
including the earlier-discussed fast Fourier transforms of the
digital samples including the extracted narrowband signal
frequencies (bins) that are harmonically related to the fundamental
frequency, the peak amplitudes of the analysis, a radio frequency
spectrum of the analysis, an accept/reject signature event
indicator, a signature part type, a signature part number, and a
manufacturer. As represented by arrows 974, 976 and block 978, a
signature correlation and selection filter is controlled to
correlate the failure signature library retained arc data with the
carrying out of arc detection and analysis prior to the analysis
for peak amplitudes as discussed in connection with block 898. It
may be recalled from FIG. 23 that comparison signatures were
initialized as described at block 940. Looking to FIG. 24,
correlation and selection as described at block 978 are further
discussed at an enhanced level of detail. The filter is entered as
represented at symbol 980 and arrow 982, leading to block 984
wherein cross correlation is carried out for signatures identified
with the index, n. As represented at arrow 986 and symbol 988, a
correlation value is saved and, as represented at arrow 990 and
symbol 992, a query is posed as to whether all signals have been
examined. In the event they have not, then as represented at
looping arrow 994, the procedure is reiterated. Where all signals
have been examined, then as represented at arrow 996 and block 998,
the procedure selects the signature with a best fit. And, as
represented at arrow 1000 and symbol 1002, the best fit signature
correlation index is saved and made available as represented in
FIG. 22 at arrow 1008 and symbol 1010. The signature ID and
correlation resultant are saved with maintenance merit data for
uploading as discussed in connection with FIG. 10.
Returning to block 898 and associated arrow 1012, the program then
carries out peak harmonic detection as represented at block 1014.
This feature has been discussed at a higher level of detail in
connection with FIG. 15. Next, as represented at arrow 1016 and
block 1018, maintenance merit computation is carried out as
described in detail in FIG. 16. The maintenance merit values then,
as represented at arrow 1020 and block 1022 are subjected to finite
impulse response filtering as represented at arrow 1020 and block
1022. The result of such filtering is represented at arrow 1024 and
symbol 1026 as a maintenance merit resultant which has been
described in detail in connection with FIG. 17. The program then
proceeds as represented at arrow 1028 and block 1030 to recording
control which, as represented at arrow 1032 enables the storage of
raw data for signal signature analysis as represented at symbol
894. Additionally, the recording control function at block 1030
carries out the features represented in FIG. 18, whereupon the
program proceeds as represented at arrow 1034 and block 1036
providing for the carrying out of arc proximity computation and
associated adjustments of the frequency response of radio 874 as
discussed in connection with FIG. 19.
Since certain changes may be made in the above apparatus and method
without departing from the scope of the disclosure herein involved,
it is intended that all matter contained in the above description
or shown in the accompanying drawings shall be interpreted as
illustrative and not in a limiting sense.
* * * * *
References